The rainbow trout is a useful lower vertebrate model for comparative issues in human environmental carcinogenesis and an appropriate surrogate to investigate the molecular basis for tumor development by feral fish populations in contaminated waterways. However, the genetic events accompanying cancer development in aquatic species are not well understood. The focus of this proposal is to characterize the molecular and cellular basis for Ki-Ras proto-oncogene activation in trout neoplasia, and its interactions with carcinogen dose, DNA adduct levels, cell proliferation and target organ toxicity down to 0.1% tumor incidence, an order of magnitude lower than any previous animal studies. We also propose a limited examination of p53 suppressor gene dominant mutation during tumor development in polyploid animals, studies not feasible with mammalian models. Specific Aims are: 1. Establish the fundamental parameters that govern the frequency, and types, of mutant Ki-ras alleles among various trout tumor phenotypes. Examine the hypothesis that the balance of ras and non-ras tumorigenesis pathways is altered by carcinogen doses (see Aim 3) or tumor promoters that induce regenerative proliferations. 2. Examine the hypothesis that selective differences in dysfunctional fitness of mutant ras p21 proteins during clonal expansion and progression, not solely site-specific rates of carcinogen-DNA adduction and mutagenesis during initiation, determine the eventual spectrum of Ki-ras mutant alleles detected in end-stage tumors. 3. Establish molecular dosimetry relationships between Ki-ras activation and carcinogen dose, DNA adduction, cell proliferation and tumor response in two ED0.1% tumor studies - By appropriate mechanistic studies explore four hypotheses: 1) that any non-linearity is explained by dose-dependent toxicity, DNA adduction, or persistence; 2) that Ki- ras oncogenesis decreases with increasing carcinogen-induced toxicity and proliferative regeneration; 3) that DMBA tumorigenesis in vivo may be driven primarily by unstable adducts from one electron oxidation, rather than stable adducts; 4) that DEN dosimetry and Ki-ras mutant allele spectrum is dependent on host excision repair proficiency, not solely alkyltransferase repair. 4. Examine the hypothesis that increased gene dosage for suppressor genes accounts for the greatly increased resistance to development of most tumor phenotypes in triploid trout. Specifically, we hypothesize that those few tumors that do develop in triploids have developed in an environment selective for dominantly transforming p53 point mutations, and thus will show a much greater prevalence of mutant p53 than will tumor from diploid siblings.